A single feed printed monopole antenna with switchable polarisation with reduced antenna size is proposed. The radiating element is in the form of two monopole rectangular patches fed by a single microstrip line. To reduce the size of the patches and to provide dual polarised operation, each patch is perturbed by two L-shaped slots. Such a structure can produce two orthogonal linear polarisations by switching only one pin diode, which is the most interesting feature of the design. To further reduce the size of the antenna a meandered stub attached to the transmission line is employed. The antenna is fabricated on the low-cost FR4 substrate with an overall size of 26 × 19 mm².

The design of an efficient rectenna with vertical ground-walls for RF energy harvesting is proposed. The proposed rectenna is composed of the broadband dipole-type antenna and the rectangular ground plane with two vertical ground-walls. Especially, the vertical ground-walls are designed to suppress and harvest the back-lobe of the main radiator without the disturbances of the main radiator's electrical performance. The back-lobe of the proposed rectenna is significantly reduced to 8.4 dB and the amount of the harvesting energy is approximately 1.74% of the original input power. This rectenna has an advantage of simple design which helps to harvest RF energy from high-gain antennas.

A frequency selective surface with a new type of three-dimensional element which produces a quasi-elliptic bandpass response is presented. The element exhibits an elliptic bandpass response with a −3 dB bandwidth (BW/f ₀) of ∼ 70%, where f ₀ is the centre frequency. The design has been adapted to a form that can be conveniently fabricated by utilising printed technology. The measured results for the frequency response show good agreement with those predicted via numerical simulations.

The performances of frequency selective surface arrays in which some of the conducting elements at randomly chosen locations were absent or defective are described. The aim was to assess the proportion of defects that can be tolerated when low-cost fabrication techniques are employed.

An innovative technique for obtaining ultra-low sidelobes pattern in a phased array is proposed. This technique involves the addition of two external edge elements which together produce an interferometer pattern used for cancellation of sidelobes. To obtain optimum cancellation of the sidelobes, the sidelobe structure of the original array must be designed to exactly match with that of the interferometer pattern.

A wideband strip-fed tapered slot antenna (TSA) for electromagnetic compatibility (EMC)/antenna measurements is presented. The proposed TSA consists of a full-shielded feeding transition from a SSMA-connector to a bilateral slotline and a pair of logarithmically tapered arms. An irregular slotline cavity is used for wideband impedance matching. The whole feeding structure is an all 50 Ω design without any multi-section impedance transformers that deteriorates the voltage standing wave ratio of the antenna. Also, the SSMA-to-strip transition is fully shielded to avoid undesired radiation loss. Experiments show that the antenna has an extremely wide operating bandwidth and can be applied to EMC/antenna measurements.

An efficient VLSI design of a lossless electrocardiogram (ECG) encoder is proposed for wireless body sensor networks. To save wireless transmission power, a novel lossless encoding algorithm had been created for ECG signal compression. The proposed algorithm consists of a novel adaptive predictor based on fuzzy decision control, and a novel hybrid entropy encoder including both a two-stage Huffman and a Golomb-Rice coding. The VLSI architecture contains only 2.71 K gate counts and its core area is 33 929 μm² synthesized by a 0.18 μm CMOS process. Moreover, this design can be operated at 100 MHz processing rate by consuming only 30 μW. It achieves an average compression rate of 2.56 for the MIT-BIH arrhythmia database. Compared with previous low-complexity and high-performance lossless ECG encoder studies, this design has a higher compression rate, lower power consumption and lower hardware cost than other VLSI designs.

An implantable slot antenna with a miniaturised size of 139.7 mm³ (10 × 11 × 1.27 mm) for biomedical applications is proposed. Operating at the Medical Implant Communications Service (MICS) band, miniaturisation of the antenna is realised by etching a meandered slot on the top metal with one end open. A microstrip line is utilised as excitation with a shorting connected to the radiator. In addition, two U-shaped stubs are adopted to match the feeding impedance to 50 Ω. The antenna is simulated in a one-layer skin model as well as in the arm of a human model and measured in skin-mimicking tissue. Good agreements can be obtained and the measurement results show that a broad bandwidth of 28.3% (from 354 to 469 MHz) for |S ₁₁| less than − 10 dB can be achieved.

Two new low-voltage micropower first-order temperature-compensated CMOS current references are presented. To achieve compact architectures able to operate under low voltage with low power consumption, the references are based on the simplest approach of cross-coupled current mirrors, and compensation is obtained by introducing a temperature-dependent current mirror ratio. Results for 0.18 µm CMOS implementations show that the proposed 1 μA references operate with supplies down to 1 V showing temperature drifts below 231 ppm/°C over a (from –40 to 120°C) range.

A low-power, high-speed full adder (FA), abbreviated as LPHS-FA, is presented as an elegant way to reduce circuit complexity and improve the performance thereof. Employing as few as 15 MOSFETs in total, an LPHS-FA requires 60–73% fewer transistors than other existing FAs with drivability. For validation purpose, HSPICE simulations are conducted on all the proposed and referenced FAs based on the TSMC 0.18-μm CMOS process technology. The LPHS-FA is found to provide a 20.4–21.2% power saving, a 12.3–67.0% delay time reduction and a 35–102% reduction in power delay product compared with the referenced FAs. In short, an LPHS-FA is presented in a concise form as a high-performance FA in practical applications.

A frequency/voltage conversion circuit for alternating current electrogravimetric measurements based on quartz crystal microbalance oscillators is proposed. This circuit improves the sensitivity and the bandwidth of the conversion systems proposed until now. Static and dynamic calibration results of the circuit are reported. These results prove its reliability for alternating current-electrogravimetric measurements and will allow the development of new applications in various electrochemical fields.

A yield formulation model to estimate the amount of lithography distortion expected in a printed layout is proposed. The yield formulation relates the probability of non-failure of a lithography hotspot with the yield loss. The application of the yield model is demonstrated for three different layout configurations showing that unidimensional designs may improve manufacturing yield.

A new type of differential transmission line with a bend is proposed to reduce differential-to-common-mode conversion noise for high-speed digital circuits. The hybrid structure consists of alternating edge-coupled and broadside-coupled transmission line sections which has the effect of reducing the mode conversion noise by at least 20 dB up to 10 GHz, whereas the differential insertion loss remains low. In addition, in the time domain, the common-mode noise is lower than the conventional edge-coupled differential transmission line bend. The performance of the proposed structure with a bend is demonstrated both numerically and experimentally. The measurement results agree very well with the simulation results from the three-dimensional full-wave solver.

Image registration is an essential and important process in seal identification. Rigid image registration in seal identification is known to be more suitable than elastic registration. The registration process is quite sensitive to outliers in matched feature point pairs. A novel method to take the matching outliers as data corrupted by ‘sample-specific’ error which can be modelled by a column sparse matrix is proposed. An optimisation problem is developed to describe this model. By solving the optimisation problem, corruption can be eliminated and the transformation model can be recovered simultaneously. An efficient algorithm called column sparse registration is given via the augmented Lagrange multiplier method. Experiments on real-world seal registration data demonstrate that the proposed method is robust to outliers among matched pairs and outperforms the state-of-the-art methods.

Among various widely used thermal non-destructive testing methods, non-stationary thermal non-destructive testing modalities have proved to be an indispensable approach for the inspection and evaluation of various solid materials. These techniques facilitate the use of low peak power heat sources in a moderate time compared with the conventional widely used pulsed and sinusoidal modulated (lock-in) thermographic techniques. In this reported work, Barker non-stationary excitation followed by Hilbert transform-based matched filtering for the detection of subsurface features lying deep inside a glass fibre-reinforced plastic test sample is incorporated. Results obtained from this pulse compression approach are less affected by random noise generated inside the test sample during experimentation as well as the variations of surface emissivity over the test sample. A comparison has been carried out among the amplitude, time delay and phase images obtained for the experimentally matched filter data.

A convex minimisation model for image segmentation is proposed. The basic idea of this model is that objects will be detected automatically if background is removed. The local information of every pixel is used to make the model applicable to images with intensity inhomogeneity. Also, by using a convex approximation of the Heaviside function, the convex energy function of the proposed model is obtained. Then it is minimised by applying the split Bregman method, which is a fast technique to obtain the global minimiser. The experimental results demonstrate that the proposed model is powerful in efficiency and accuracy.

An efficient contextual characteristic is proposed for person re-identification. Most current approaches are based on either constructing robust appearance descriptors or learning a distance metric for precise feature matching. However, re-identifying results may be inaccurate and not robust due to appearance features variation caused by various environment changes and individual movement factors. In this reported work consideration is given to the introduction of the contextual characteristic that contains similarities of both k-nearest and ḱ-farthest neighbours between the probe and the gallery, and combines it with Mahalanobis distance for ranking every gallery image more accurately. The experimental result has validated the effectiveness of the proposed method on a challenging publicly available dataset.

A new adaptation mode control method using the multiple signal classification (MUSIC) algorithm is proposed. The MUSIC algorithm is used to calculate the spatial spectral response to the direction of the target source with noisy inputs. Then, the statistical distribution of the MUSIC output is analysed and used to model the target signal presence probability. Using the probability, the filter adaptation mode of a multiple input canceller in a generalised sidelobe canceller is controlled. For performance evaluation, hidden Markov model-based speech recognition rates are measured. The results show that the proposed algorithm notably outperforms other conventional adaptation mode controllers.

Complex networks play a significant role in modern complex systems sciences in that they allow for the quantitative analysis of the structural properties of systems composed of different interacting entities. Recently, intensive efforts have been made to apply network-based concepts also for the analysis of time series. In this reported work, multiscale complex networks from the chaotic time series are constructed by a decomposition strategy. Since different components exist in the time series, it was found that the constructed network inherits the multiscale properties of the time series in its structure. For example, periodic series and noisy series convert into regular networks and random networks, respectively. A novel method for the chaotic time series analysis is provided.

A novel management scheme for the write buffer in solid-state drives (SSDs) is presented. The proposed scheme exploits the future buffer reference pattern by using I/O commands information in native command queuing (NCQ) of SATA SSDs. Through the trace-driven simulations, it is shown that the proposed scheme improves the performance of the write buffer significantly in terms of several metrics including the hit ratio.

A switching-based variable step-size approach for partial update least-mean-square (LMS) algorithms is proposed. During the initial stage, the errors are dominated by the mismatch between the tap weights of the adaptive filter and its ideal weights. In this reported work, the error signals are correlated with the linear interpolated error signals to gain a large step-size; on the other hand, during the steady state, the errors mainly come from the additive noise. Therefore, the correlation is switched to the other mode so that the effect of noise can be eliminated. This can be done by correlating the error signals with a delayed version of the error signals, hence a small step-size is obtained during the late stages. Simulation results show that when only a half of all taps are updated in one iteration, the proposed method significantly enhances the convergence rate, especially for the ‘data independent’ partial update LMS algorithms.

A novel dual-band bandpass filter (BPF) with a dual-plane structure is proposed. The dual-plane structure consists of a pair of microstrip feed lines with capacitive source–load coupling on the top layer and a meander multimode defected ground structure resonator (MMDGSR) on the bottom layer. Using the intrinsic characteristics of the MMDGSR, two passbands and four transmission zeros can be generated simultaneously. An experimental dual-band BPF operating at 2.45/5.8 GHz (WLAN application) has been designed and fabricated. This performance is verified by good agreement between the simulated and measured results.

New frequency-variant losses of planar thin film transmission lines are experimentally investigated in a broad frequency range. The frequency-variant transmission line parameters are accurately determined in the measured frequency band (i.e. from 40 MHz to 50 GHz). Then, it is shown that there are three critical frequencies that characterise the loss mechanism of thin film transmission lines. The conventional skin-effect model is not accurate in thin and fine transmission lines.

Graphene field-effect transistors (GFETs) are characterised for the first time under electrostatic discharge stresses. The GFETs are measured from the transmission line pulsing (TLP) tester and very fast TLP (VFTLP) tester. The turn-on behaviour influenced by back gate voltage is investigated. The I–V curve of the GFETs shows no characteristic of snapback from TLP or VFTLP measurement.

An efficient optical comb generation method based on degenerated four-wave-mixing process in highly nonlinear fibre is demonstrated. The all-optical processing of the frequency chirp is wavelength tunable and thus enables the capability of tunable chirp magnification. The magnification of the frequency chirp with a factor tunable from two to five has been demonstrated. An optical comb with 22 sub-carriers has been obtained which refers to a bandwidth of 210 GHz, about five times the bandwidth of 40 GHz for the original signal.

The theory, design, implementation and experimental validation of a hardware-circuit-based diagnosis method for open-switch faults in voltage source inverters are discussed. Operating states of the inverter under both healthy and faulty conditions are analysed. The estimated voltages can be achieved as pre-fault voltages by means of the switching function model of the inverter under a normal condition. Then, the error voltages between the estimated ones and measured ones can be used to detect faults. To achieve reliable diagnosis, the design employs both the amplitude limit and the duration limit to filter the error voltages so as to eliminate effects of the deadtimes and the switching processes of power devices. The proposed method is validated by experiments.

Domestic induction cookers can have several heating areas, hence multiple-output topologies are commonly used. One of these multiple-output topologies is the double series-resonant half-bridge inverter. An upgrading of this topology is proposed. This upgrading consists of adding a relay to allow connecting the capacitor tanks in parallel. The aim of this upgrading is to increase the efficiency through the reduction of the device currents. The proposed upgrading topology has been tested by means of a prototype.

A virtual Hall method for sensorless control of interior permanent magnet synchronous motors (IPMSMs) using high-frequency signal injection is presented. Based on the idea of virtual Hall, the position signal is easily estimated by comparing the magnitudes of three-phase inductances. The experimental results prove that the proposed method has good heavy-load operation capability with a brief control structure.

First, a polarimetric three-scale scattering model, including a large-scale-slope-induced polarisation orientation angle shift, a moderate-scale orientation angle disturbance and a small-scale Bragg resonance, to describe the polarimetric coherence matrix for a sloped rough surface is introduced. Then, a look-up table is created to demonstrate the relations between polarimetric scattering coefficients and the dielectric constant varying with different terrain slopes. Finally, a look-up-table-based inversion method is developed and airborne AgriSAR 2006 data for a sloped bare soil area are chosen to validate the inversion.

A dual-band superconducting microstrip filter with two passbands of 2.985–3.015 GHz and 3.045–3.075 GHz, respectively, is synthesised by a frequency transformation method. Open-loop resonators with embedded single spiral resonators are proposed to realise the synthesised resonator frequencies and coupling coefficients. This structure has the advantages of compact size and weak parasitic couplings, which benefit the implementation of the dual-band planar filter. The filter is successfully designed and fabricated on an MgO substrate. The measured results show high performance and match well with the simulations.

One of the challenges to implementing downlink multiuser multiple-input multiple-output in practical systems is to overcome the detrimental impact of the imperfect channel state information at transmitter (CSIT). Researchers have derived a low bound to describe the imperfect CSIT's impact on the achievable rate. However, this low bound is not tight, especially when the quality of CSIT is poor. A tighter low bound using the exponential integral is proposed. A strict proof is given and the simulation results verify the tightness.

An adaptive sensing period optimisation scheme for cognitive radio networks based on a multi-objective genetic algorithm formulation is proposed. The proposed scheme aims at maximising the spectrum opportunities as well as minimising the incurred sensing overhead. The simulation results show that the proposed scheme outperforms the non-optimised proposals by up to 90%. It can also obtain similar or superior performance to the compared schemes described in the literature, while keeping the sensing overhead within a target value.

Presented is a new compact green design structure of distributed amplifiers in series with a common-source circuit to reduce power-dissipation and noise figure and increase gain-bandwidth. Fabricated in a 90 nm CMOS standard process, the integrated circuit prototype achieves a unity-gain cutoff frequency of 42 GHz and a gain of 8.4 dB, with a gain flatness of ± 1 dB over the bandwidth of 10–30 GHz. The noise figures are between 3.9 and 5.2 dB from 10 to 30 GHz of power consumption, and the current of the distributed amplifier is 11.13 mW, and 18.5 mA at 0.6 V supply voltage, respectively. The analytical, simulated and measured results are mutually consistent.

An optimal beamforming design for spectrum sharing cognitive non-regenerative multi-relay networks that maximises the signal-to-noise ratio of the secondary user subject to the total transmit power constraint at the relays and the interference power constraint at the primary user is proposed. In the power-limited regime and the interference-limited regime, closed-form beamformers can be found. While in the intermediate operating regime, the beamforming problem is first transformed to a convex semidefinite programming (SDP) via rank-one relaxation and Charnes-Cooper transformation, and then the optimal beamformers are found by applying the rank-one decomposition theorem to the solution of the established SDP. Simulation results show that the proposed solution performs better than existing schemes.

Conversion models, originally devised to turn yearly rainfall statistics from long (e.g. 30 or 60 min) to short integration time T (i.e. 1 min), are assessed for their ability to also predict monthly 1-min integrated statistics, P(R)₁ ^m , knowledge of which may be beneficial for specific services (e.g. reconfigurable systems) and for the definition of a reliable approach to estimate monthly (hence worst month) rain attenuation statistics. Tests, performed for 5 ≤ T ≤ 60 min against monthly raingauge-derived rainfall data collected in some sites worldwide, indicate that the EXponential CELL rainfall statistics conversion (EXCELL RSC) and Lavergnat-Golé models, in force of their physical soundness, provide a good performance when used to predict 1-min integrated rainfall statistics both on yearly and on monthly bases.

Localisation error due to sensitivity to position ambiguity, with respect to some neighbouring nodes, is a major problem in wireless localisation. To solve this problem, the tangent linear approximation (TLA) for the nonlinear localisation model is derived, contributing to incorporating the effect of the neighbours’ position ambiguity into belief propagation (BP) localisation. Theoretical analysis and numerical results indicate that the TLA-BP algorithm can acquire higher accuracy and convergence probability compared with other typical methods.

A novel cyclostationarity-detection-based simple spectrum sensing technique of orthogonal frequency division multiplexing signals for cognitive radio networks is presented. The presented techniques do not require any reference tables to compute a cyclic autocorrelation function. The presented results are compared with the conventional results. Although these are found to be almost equal, the presented techniques have a lower hardware cost than the conventional techniques.